Better Betas

We provide a data-driven adjustment for estimated betas that leads to material improvements in the accuracy of weights and risk forecasts for minimum variance portfolios. Like the widely used Blume 2/3 rule and Vasicek correction developed in the 1970s, our beta adjustment operates by shrinking raw beta estimates toward 1. Unlike its antecedents, our adjustment adapts dynamically to volatility regimes, which matters since betas have tended to become more dispersed in stressed markets. Our analysis underscores the negative relationship between beta concentration and diversification, forcing a re-examination of the rule of thumb that diversification disappears in a crisis